Locally Estimated Scatterplot Smoothing for Censored Survival Data
David van Klaveren, Peter C. Austin, Patrick W. Serruys, Ben van Calster, Frank E. Harrell · 2026
Read Locally Estimated Scatterplot Smoothing for Censored Survival Data by David van Klaveren, Peter C. Austin, Patrick W. Serruys, Ben van Calster, Frank E. Harrell (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13675719.txt.
Description
Van Klaveren and co-authors present a statistics method, published in 2026, for drawing smooth risk curves when follow-up data are cut short. Doctors often check whether a prediction model is well calibrated by plotting event rates against a risk factor. The standard smoothing tool for such plots assumes every patient was followed to the end. In real studies, many patients drop out early, which bends the curve. The team adapted the smoother to handle that censoring.
Their fix weights each patient by the chance they stayed under observation. Those weights come from survival methods, with a flexible model when dropout itself depends on the risk factor. The weighted smoother then recovers the true shape of the risk relation. Tests in the paper check the method's accuracy. Calibrated models are the base of fair medical prediction, so the fix has wide use.
Catalog record
| CREATOR | David van Klaveren, Peter C. Austin, Patrick W. Serruys, Ben van Calster, Frank E. Harrell |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 31 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |